Biomarker-based early detection can directly advance prevention research by identifying key molecular events that drive cancer initiation, progression, and outcomes. Few biomarkers exist for the early detection or progression of non-Hodgkin lymphoma (NHL), a cancer of immune cells which has experienced one of the largest and still unexplained - increases in incidence and for which more than 66,000 new cases are diagnosed annually. Two promising areas of NHL research for identifying biomarkers of early detection and prognosis are (i.) host genetics and (ii) molecularly-defined tumor subtypes. Our study objective is to evaluate the role of host genetic variations and tumor molecular subtypes in NHL survival among over 1000 females diagnosed with NHL (2004-2008) who were enrolled in the Los Angeles (LA) County NHL Case-Control study. Consortial efforts now clearly implicate human leukocyte antigen (HLA) Class I and Class II genes and the tumor necrosis factor (TNF) gene in NHL etiology. Clinical studies further suggest associations between HLA and TNF in NHL survival. In exploratory analyses, we have further identified associations with NHL survival with HLA-DRB1*13 and HLA-Bw4, HLA alleles notably associated with established NHL risk factors (e.g., viral infections, acquired immunodeficiency syndrome). We believe detailed investigation into the role of immune genes with a particular focus on HLA in NHL survival is warranted to follow-up these intriguing results. In Aim 1 we will investigate the prognostic significance of genetic variation in the human leukocyte antigen (HLA), and other key a priori genes linked to HLA, including TNF, in overall survival. While host genetics may reflect a general immune milieu that affects disease progression, molecular characteristics within the tumor are thought to reflect the disease process itself. A series of landmark publications based on gene expression data has yielded molecularly defined NHL subtypes that reflect different survival. A number of tumor marker algorithms based on immunohistochemical (IHC) staining have been developed to simulate these molecular subtypes but attempts to replicate IHC-based algorithms have been mixed. We thus propose a systematic evaluation of these algorithms to move this translational research forward. In Aim 2, we will evaluate the prognostic significance of a priori and emerging tumor markers in overall survival, with a focus on the two most common NHL subtypes, diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma. Our study is among only a handful with NHL cases treated in the immunotherapy (rituximab) era (post-2000), making results generalizable to current patient populations. Our ability to successfully retrieve tumor tissue and obtain detailed therapy information will inform the predictive nature of host genetics and molecular tumor markers in NHL survival.